Abstract:Weakly supervised object detection (WSOD) and semantic segmentation with image-level annotations have attracted extensive attention due to their high label efficiency. Multiple instance learning (MIL) offers a feasible solution for the two tasks by treating each image as a bag with a series of instances (object regions or pixels) and identifying foreground instances that contribute to bag classification. However, conventional MIL paradigms often suffer from issues, e.g., discriminative instance domination and missing instances. In this article, we observe that negative instances usually contain valuable deterministic information, which is the key to solving the two issues. Motivated by this, we propose a novel MIL paradigm based on negative deterministic information (NDI), termed NDI-MIL, which is based on two core designs with a progressive relation: NDI collection and negative contrastive learning (NCL). In NDI collection, we identify and distill NDI from negative instances online by a dynamic feature bank. The collected NDI is then utilized in a NCL mechanism to locate and punish those discriminative regions, by which the discriminative instance domination and missing instances issues are effectively addressed, leading to improved object-and pixel-level localization accuracy and completeness. In addition, we design an NDI-guided instance selection (NGIS) strategy to further enhance the systematic performance. Experimental results on several public benchmarks, including PASCAL VOC 2007, PASCAL VOC 2012, and MS COCO, show that our method achieves satisfactory performance. The code is available at: https://github.com/GC-WSL/NDI.

Multi-instance Learning Based on Instance Consistency for Image Retrieval

Semantic Image Retrieval Based on Multiple-Instance Learning

Multiple-Instance Learning from Pairwise Comparison Bags

Web Image Clustering Based-on Multiple Instance Learning

Multiple instance learning from similarity-confidence bags

Image Annotation by Multiple-Instance Learning with Discriminative Feature Mapping and Selection

Concurrent Multiple Instance Learning For Image Categorization

Multi-instance Positive and Unlabeled Learning with Bi-Level Embedding

Multiple-Instance Learning from Similar and Dissimilar Bags

Interactive Localized Content Based Image Retrieval with Multiple-Instance Active Learning

Localized Content-Based Image Retrieval Using Semi-Supervised Multiple Instance Learning

Text-based image retrieval using progressive multi-instance learning

Multi-Instance Learning with Key Instance Shift

Multiple-Instance Learning from Unlabeled Bags with Pairwise Similarity

Multiple-instance learning with instance selection via constructive covering algorithm

A New Multi-Instance Learning Scheme For Scene Categorization Using Information Bottleneck Theory

Scalable Algorithms for Multi-Instance Learning.

Region-Based Image Annotation Using Heuristic Support Vector Machine in Multiple-Instance Learning

Negative Deterministic Information-Based Multiple Instance Learning for Weakly Supervised Object Detection and Segmentation

Instance-level Semisupervised Multiple Instance Learning.

Learning With Only Multiple Instance Positive Bags